Journal of Hazardous Materials B128 (2006) 3943

Extraction and recovery of chromium from electroplating sludge

Paula Tereza de Souza e Silva a, , Nielson Torres de Mello b , Marta Maria Menezes Duarte c , M. Conceicao B.S.M. Montenegro d , d , Bencio de Barros Neto a , Valdinete Lins da Silva b Alberto N. Ara jo u
b

Departamento de Qumica Fundamental, Universidade Federal de Pernambuco, Cidade Universit ria, 50740-250 Recife-PE, Brazil a Departamento de Engenharia Qumica, Universidade Federal de Pernambuco, Rua Teresa Melia, s/n Cidade Universit ria, 50740-250 Recife-PE, Brazil a c Funda ao do Instituto Tecnol gico do Estado de Pernambuco-ITEP, Av. Prof. Luiz Freire, 700 Cidade Universit ria, 50740-540 Recife-PE, Brazil c o a d REQUIMTE/Dep. Qumica-Fsica, Faculdade de Farm cia, Rua Anbal Cunha 164, 4099-030 Porto, Portugal a Received 7 April 2005; received in revised form 15 July 2005; accepted 16 July 2005 Available online 6 September 2005

Abstract This work reports a study of the extraction and recovery of chromium from the wastes (class I dangerous) generated by a galvanic manufacturer. Commercial HCl at room temperature was employed, and the conditions of the extraction process were optimized according to a sequential experimental design, which also included the acid concentration and contact time as variables. The best extraction conditions (80% v/v; 30 min; 97.6% Cr) for the chromic sludge were chosen in order to make the recovery process economically feasible. After each extraction, the residue was submitted to leaching essays, to assess environmental risks. It was found that sludge could be characterized as no longer dangerous. In the recovery study, a simple and low-cost technique was evaluated for selectivity based on an oxidation step with hydrogen peroxide. A 23 factorial design to assess the inuence of oxidation time (min), temperature ( C) and peroxide amount (mol/L) was employed. The best conditions, yielding a chromium recovery of about 92%, were a time of 60 min, a temperature of 60 C and 2.1 mol/L peroxide. Additional essays revealed that the same result could be obtained with more economic conditions (40 min, 1.4 mol/L peroxide and 60 C). This technique proved not only effective in comparison with existing alternatives, but also low costing. 2005 Elsevier B.V. All rights reserved.
Keywords: Galvanic manufacturing; Wastes; Chromium; Extraction; Recovery

1. Introduction Residues discarded by industrial processes gure among the most important sources of environmental contamination. This is especially true of processes that use or yield dangerous chemicals, leading to potentially dangerous residues [1]. Many noxious residues are routinely treated for removal and recovery of heavy metals, with techniques that include conventional neutralization or precipitation, electrochemical reduction, ion exchange, reverse osmosis, membrane separation, and solvent extraction [24]. Precipitation of metal ions as hydroxides, under appropriate pH conditions, has been one
Corresponding author. Tel.: +55 81 21267290; fax: +55 81 34232547. E-mail addresses: paulatss@yahoo.com.br (P.T. de Souza e Silva), vlins@yahoo.com.br (V.L. da Silva). 0304-3894/$ see front matter 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.jhazmat.2005.07.026

of the most used conventional methods [5]. This treatment, however, generates a heavy metal laden slime or sludge. The literature reports only few works on heavy metal recycling, and is scarce concerning chromium found in sludge resulting from chemical neutralization and precipitation. Since recovery of these metals with traditional separation procedures does not appear to be economically feasible, a search for safe alternative solutions for their nal disposition is warranted [6]. One such possibility is using the metal sludge as an additive to cements and ceramic materials [7,8]. The main drawback of this approach resides in the fact that many people are allergic to heavy metals, and may exhibit skin edemas because of contact with the cement wet paste or with the ceramic artifact itself. Moreover, in addition to their toxic potential some sludge metals are economically valuable. This lends further motivation for investigating other recovery pro-

40

P.T. de Souza e Silva et al. / Journal of Hazardous Materials B128 (2006) 3943 Table 1 Sequential factorial designs used to evaluate the inuence of acid concentration and contact time on heavy metal extraction Run 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 HCl concentration (% v/v) 60 80 100 60 80 100 60 80 100 30 60 80 100 30 60 80 100 10 10 Contact time 24 h 24 h 24 h 5h 5h 5h 3h 3h 3h 1h 1h 1h 1h 30 min 30 min 30 min 30 min 10 min 30 min

cesses. Some hydrometallurgical techniques to treat sludge have been studied, with a view to obtaining maximum selectivity and recycling. Promising results have been obtained by combining relatively difcult and expensive methods like conventional liquidliquid extraction, ion exchange, and/or electrochemical separations. These attempts, however, have not advanced beyond the laboratory or pilot plant scale [9,10]. In this work, we report on the treatment of galvanic sludge residues originating from a pre-treatment step in a production line for coated aluminum plates. This process consists of ve stages: (1) alkaline degreasing at pH 2; (2) alkaline rinsing at pH 8.59.0; (3) chromium electrodeposition in a uoridric/phosphoric medium; (4 and 5) two successive rinsing steps with industrial water at pH 3.0 and 5.0, respectively. After this treatment, the plates are ready to be taken for the coil coating step. All baths are carried out by spraying on both faces of the plates, with a nozzle pressure of 2 Kgf/cm2 and a line speed of 1730 m/min, depending on plate thickness. The temperature is maintained at 35 1 C throughout the whole process. The efuent from the galvanic process is then neutralized and the metals precipitated by adding sodium hydroxide. The resulting sludge is nally treated by a simple, efcient and low-cost methodology for chromium extraction and recovery, evaluated in this work. This methods parameters were systematically varied according to sequential twolevel factorial designs, to improve its extraction efciency and to evaluate the selectivity of chromium oxidation by hydrogen peroxide. This methodology was preferred, instead of the widely used simplex optimization, because it usually requires less experimental runs.

2.3. Acid extraction of metals from the sludge Low-costing commercial HCl (Laborqumica) (US$ 0.6/L) was used as the extracting agent, with the following characteristics: 25.8% purity, 1.13 g/mL density, 0.02 mg/L Zn, 0.03 mg/L Cr, 0.06 mg/L Al. For metal extraction, 5 mL HCl was added to a residue sample weighing approximately 1 g, which was then stirred at 150 rpm. To nd out improved extraction conditions, the inuence of acid concentration and contact time was assessed with the four factorial designs given in Table 1, carried out sequentially. After each experiment, the heavy metal amounts were determined by ICP-OES. The residue left after each extraction was again subjected to leaching [11]. 2.4. Chromium recovery To assess chromium recovery, Cr(III) was oxidized to Cr(VI) with 30% hydrogen peroxide (US$ 1.5), according to 2CrO2 + 3H2 O2 2CrO4 2 + OH + H2 O (1)

2. Experimental 2.1. Classication of chromic sludge A chromic sludge gross sample obtained from treatment of galvanic industrial efuents was subjected to leaching and solubilization, as previously described [11,12]. Chloride, cyanide, total chromium, uoride, arsenic, barium, bismuth, chromium, mercury, nitrate, total phosphorus, sulfate, and surfactants were then quantied, following the methodology of [13]. 2.2. Chemical composition of the chromic sludge An amount of approximately 1 g of sludge sample was weighed and transferred to a porcelain crucible, mixed with 5 mL of concentrated HNO3 , and digested on the top of a heating plate. This procedure was repeated until the supernatant was clear. The material was then ltered into a 100 mL volumetric ask, and the volume completed with distilled water. A blank was prepared in the same way. The metals were quantied by inductively coupled optical emission spectrometry (ICP-OES).

A real sludge sample containing Cr (2424.5 mg/L), Al (1625.0 mg/L), Fe (1220.6 mg/L), Ca (157.1 mg/L), and Mg (155.3 mg/L) was employed for the analysis. The 23 factorial design shown in Table 2 [14] was employed to achieve improved recovery conditions. The three factors in this design concentration of H2 O2 , reaction time and temperature were selected according to studies found in the literature [15,16]. The percent of oxidized Cr(III) was used as the response in the statistical analysis of the design results. The analysis of this rst design indicated the need for the additional experiments given in the last ve lines of Table 2. All

P.T. de Souza e Silva et al. / Journal of Hazardous Materials B128 (2006) 3943 Table 2 Factorial designs used for studying Cr(III) oxidation conditions Run 1 2 3 4 5 6 7 8 911 12 13 14 15 16 Time (min) 20 60 20 60 20 60 20 60 40 40 40 60 60 60 Temperature ( C) 40 40 60 60 40 40 60 60 50 60 60 50 50 60 [H2 O2 ] (mol/L) 0.7 0.7 0.7 0.7 2.1 2.1 2.1 2.1 1.4 1.4 2.1 1.4 2.1 1.4 Cr(VI) (mg/L) 205.84 931.98 1421.48 1853.78 385.25 1894.75 1666.36 2253.57 1372.27, 1184.85, 1372.76 2253.57 2123.14 1772.31 1829.53 2172.35

41

Oxidized Cr(III) to Cr(VI) (%) 8.49 38.44 58.63 76.46 15.89 78.15 68.73 92.95 56.60, 48.87, 56.61 92.95 87.57 73.10 75.46 89.60

The rst design included a central point in triplicate (runs 911).

calculations were carried out with the Statistica computational package [17]. 2.5. Procedure for oxidizing Cr(III) to Cr(VI) A 20 mL aliquot of the extract containing 2424.5 mg/L chromium was transferred to a 50 mL beaker and its pH was adjusted to 10 with NaOH, as indicated by the results obtained by [18,19]. The experiments were performed as specied by the level combinations given in Table 2. After each experimental run, the corresponding solution was cooled and ltered. The concentration of chromium(VI) in the ltrate was determined by the diphenylcarbazide method, using an UVvis spectrophotometer, with absorbances recorded at 540 nm [13].

3.2. Sequential experimental design for the chromic sludge Several studies with sludge resulting from the treatment of efuents have been reported. Within these, Macchi et al. obtained 90% chromium recovery using concentrated H2 SO4 and a contact time of 24 h [18]. Taking these results as starting point, in the rst experiments the contact time was kept at 24 h and the concentration of commercial hydrochloric acid (used instead of the more expensive sulfuric acid) was set at three different levels: 60, 80 and 100% (v/v). These experiments led to chromium recoveries in the 89.999.6% range. Moreover, almost all the residue was dissolved, rendering the uoride leaching determination unnecessary. With a view to reducing process costs, shorter contact times (5 and 3 h) were then tried. Chromium recovery in these experiments remained high, from 87.5 to 99.7%, and again practically all the sludge dissolved. In the third design, the acid concentration was varied from 30 to 100%, and the contact time from 30 min to 1 h. Recovery values between 75.1 and 99.5% were then obtained. The concentration of leached uoride was determined only for recovery values under 81.0%, where it was found in the 30.636.1 mg/L interval, well within the range allowed by the regulation [1]. In the last design, the acid concentration of acid was reduced to 10% and the contact time was varied from 10 to 30 min. These levels yielded recoveries from 55.5 to 61.8%. Even though, in the cases when a residue was left, the largest leached uoride concentration was 47.7 mg/L, still in compliance with legislation. These relatively low amounts may be explained by the low solubility of Cr(III) in water. A 3D plot of the results of all experiments is shown in Fig. 1. As the acid concentration and the contact time increase, the amount of recovered chromium also increases. This may be explained by an increase in dissolving power as the acid concentration is raised. This is an important factor, because Cr(III) is insoluble in water. Longer contact times also lead to increased

3. Results and discussion 3.1. Characterization of chromic sludge The chemical composition of chromic sludge is determined by the process used for treating industrial efuents, chromium, aluminum, iron, calcium, and magnesium being the elements present in the largest amounts. The usual treatment consists in using sodium bisulte to reduce the Cr(VI) ions resulting from the painting process of aluminum plates to Cr(III), after which the highly insoluble chromium(III) hydroxide is precipitated with NaOH. Several other metals also precipitate in this step from the efuent, where they are present in the following amounts: Fe (1850.0 mg/L), Al (1891.7 mg/L), Ca (225.1 mg/L), Mg (212.9 mg/L), Mn (1.8 mg/L), and Zn (5.0 mg/L). After the chromic sludge was subjected to leaching according to the [11] procedure, it was found that uoride was still present in a concentration of 249.0 mg/L, well above the maximum level allowed by legislation, 150.0 mg/L. Therefore, the residue is classied as dangerous, class 1 [1].

42

P.T. de Souza e Silva et al. / Journal of Hazardous Materials B128 (2006) 3943 Table 3 Analysis of the 23 design corresponding to runs 111 in Table 2 Main Effects t 33.6 T 38.9 [H2 O2 ] 18.4 Interactions tT 12.5 t[H2 O2 ] 9.71 T[H2 O2 ] 5.1 tT[H2 O2 ] 6.5

All effects have a standard error of 3.1%. Statistically signicant values are shown in boldface.

Fig. 1. Percent chromium recovery in chromic sludge at several concentrations of commercial hydrochloric acid (5, 10, 30, 60, 80, and 100%) and contact times (10 min 0.17 h, 30 min 0.5, 1, 3, 5, and 24 h). = =

recovery, but this effect is not very signicant for concentrated solutions. 3.3. Chromium recovery from the sludge Since chromium is not the sole metal ion present in the sludge, hydrogen peroxide was employed to oxidize Cr(III) to Cr(VI), to avoid interference of other species in its recovery. Chromium in sludge usually appears as part of insoluble hydroxides. To separate it from other metals, several oxidizing agents have been used. Mathew used sodium hypochlorite, ozone and hydrogen peroxide to oxidize chromium in electroplating sludge [19]. Even with excess sodium hypochlorite, chromium oxidation required a medium set at pH 10, heating to 50 C and stirring for 1 h, and no more than 85% chromium was recovered as chromate. Duffey obtained complete Cr(III) recovery from an iron-containing sludge, also using sodium hypochlorite [20]. It was suggested, in this case, that the ferric ion remains insoluble as ferric hydroxide, instead of ferric chromate, and therefore does not compete with chromium. Despite being a very efcient oxidizing agent for Cr(III), sodium hypochlorite may exhibit some undesirable environmental effects. During the oxidation of Cr(III), the organic matter in the sludge may react with chlorine in the hypochlorite to yield harmful organochlorinated compounds [18,21]. Human exposure to these compounds may

lead to headaches, skin thickening, haemorrhage, testes atrophy, embrionary alterations, and kidney cancer [22]. Ozone is a stronger oxidizing agent than hypochlorite is, but its production cost is considerably higher. Mathew [19] used ozone to oxidize Cr(III) in sludge, but obtained only 67% recovery thus concluding that using stronger oxidants did not necessary lead to increased recoveries. The effects calculated for the 23 design corresponding to runs 111 in Table 2 are shown in Table 3. Only the main effects were found signicant. They are all positive, indicating that higher recovery values will be obtained when all three design factors are set at their higher levels. Complete chromium oxidation was not achieved. The maximum recovery was 92.5%, which is comparable to the 84% value reported by [19]. These authors argued that this limitation might be due a polymer formed by Cr(III) and Cr O Cr, and/or to Cr OH Cr bridges, or even to the presence of iron, which may compete with chromium and preclude the latters oxidation. According to [23], oxidation of these polymers by hydrogen peroxide is hard to achieve. In an attempt to nd more economical recovery conditions, we decided to perform additional runs at level combinations beyond those specied by the starting 23 design, and displaced toward the region associated with higher recovery values. The results of these runs are shown in the last ve lines of Table 2. In these new experiments, a maximum of 92% chromium recovery was obtained with run 12, which corresponds to the settings of t = 40 min, T = 60 C, and [H2 O2 ] = 1.4 mol/L. The same value was obtained with the best run of the starting design (no. 8), which, though carried out also at 60 C, required more time (60 min) and peroxide concentration (2.1 mol/L). This probably means that at the higher temperature the factors contact time and [H2 O2 ] become relatively secondary, and can be used at reduced settings, without signicantly affecting chromium recovery. Since most metals precipitate as hydroxides, all metal amounts found after the oxidative process (Zn, Mn, Fe, Al, Ca, and Mg) were very low, and did not compromise the purity of the process solution (Table 4). Residual iron was very easily removed at pH 710. Even with the increased

Table 4 Chemical composition of the sludge solution, before and after the oxidative process (mg/L) Metal Before After Zn 5.0 0.005 Mn 1.4 0.001 Fe 1220 0.035 Cr 2424 2287 Al 1625 5.6 Ca 157 1.8 Mg 155 0.025

P.T. de Souza e Silva et al. / Journal of Hazardous Materials B128 (2006) 3943

43

solubility of aluminum at pH 10 as Al(OH)4 , most of the aluminum was removed. All precipitated metals were separated by ltration.

[7]

4. Conclusions Sequential designs were used to nd satisfactory experimental conditions for chromium recovery from electroplating sludge. A recovery of 97.6% was attained with 30 min contact time and 80% (v/v) HCl (US$ 0.6/L). In every case the leached uoride levels were high enough to characterize the residuals as dangerous, class 1. Hydrogen peroxide (US$ 0.9/L) was employed at 30% because of its fast kinetics in relation to O2 [24], and because it decomposes to oxygen and water, which are not environmentally damaging [18]. With the oxidative process it was possible to attain a 92% recovery of metallurgical grade chromium at 40 min, 60 C and 1.4 mol/L. The methodology reported here is simple, economical and environmentally sound, since it requires only low-cost reagents and little energy.

[8]

[9] [10]

[11] [12] [13]

[14]

[15]

Acknowledgments Financial and material support from the funding agencies FACEPE, CAPES, and CNPq, and also from ITEP e ALCOA are gratefully acknowledged.

[16]

[17] [18]

References
[1] ASSOCIACAO BRASILEIRA DE NORMAS TECNICAS-ABNT, NBR 10.004-Resduos. Rio de Janeiro, 1987. [2] R.A. Corbitt, Standard Handbook of Environmental Engineering, McGraw-Hill, New York, 1990. [3] R.C. Wilmoth, S.J. Hubbard, J.O. Burckle, J.F. Martin, Occurrence, Analysis and Biological Relevance, VCH, Weinheim, Germany, 1991, pp. 1965. [4] M. Henze, P. Harremoes, J. Jansen, E. Arvine, WasteWater Treatment, Springer-Verlag, Berlin, 1995. [5] C.A.L. Chernicharo, M. von Speling, A new approach for ambiental control industry, in: Proceedings of International Seminary of Tendencies in Simplied Treatment of Domestic and Industrial WasterWater, Brazil, 1996, pp. 158166. [6] B. Stoll, A.R. Bidone, J. Povinelli, Caracterizacao Fsico-Qumica de lixiviados de serragens e aparas de couros dispostas em c lulas e

[19]

[20]

[21]

[22] [23]

[24]

de aterros sanit rios, in: 20 Congresso Brasileiro de Engenharia a Sanit ria Ambiental, 1999, pp. 20452054. a A. Borges, R. Soares, V. Amaral, Solidicacao/Estabilizacao de Lodos Galv nicos (Metais: cromo e zinco) em matrizes de cimento a portland, in: 20 Congresso Brasileiro de Engenharia Sanit ria Ambia ental, 1999, pp. 19. W. Keller, C. Chernicharo, J. Lauffer, R. Moreira, O cromo nos resduos s lidos de curtumes imobilizacao em materiais o cer micos, in: 20 Congresso Brasileiro de Engenharia Sanit ria a a Ambiental, 1999, pp. 17591768. D.S. Jadhar, Z.R.J. Turel, Radioanal. Nucl. Chem 177 (1994) 185. J. Jandov , T. Stefanov , R. Niemezykov , Recovery of Cu cona a a centrates from waste galvanic copper sludges, Hydrometallurgy 57 (2000) 7784. ASSOCIACAO BRASILEIRA DE NORMAS TECNICAS-ABNT, NBR 10.005-Lixiviacao de resduos, Rio de Janeiro, 1987. ASSOCIACAO BRASILEIRA DE NORMAS TECNICAS-ABNT, NBR 10.006-Solubilizacao de resduos, Rio de Janeiro, 1987. Standard Methods for the Examination of Water and Wastewater, 20th ed., American Public Health Association/American Water Works Association/Water Environmental Federation, Washington, DC, USA, 1998. B.B. Neto, I.S. Scarminio, R.E. Bruns, Como fazer experimentos, in: Pesquisa e Desenvolvimento na Ci ncia e na Ind stria, UNICAMP, e u Campinas, SP, 2001. W.J. Cooper, R.G. Zika, Photochemical formation of hydrogen peroxide in surface and groundwaters exposed to sunlight, Science 220 (1983) 711712. R.G. Zika, E. Saltzman, W.J. Cooper, Photochemical formation of hydrogen peroxide in surface and groundwaters exposed to sunlight, Mar. Chem. 17 (1985) 265275. Statsoft Inc., Statistica for windows [Computer program manual], 1999. WEB: http://www.statsoft.com. G. Macchi, M. Pagano, M. Pettine, M. Santori, G. Tiravanti, A Bench study on chromium recovery from tannery sludge, Water Res. 25 (1991) 10191026. G. Matthew, J. Beattie, B. Kennedy, Recovery of chromate from electroplating sludge, Waste Manage. Res. 18 (2000) 380 385. J.G. Duffey, Chrome recovery from ferric chromic hydroxide by oxidation to dichromate, in: Consumer and Corporate Affairs Canada, Environmental Processes Inc., Canada, 1981. F.K.M. Kazi, P.A. Cooper, Rapid-extraction oxidation process to recover and reuse copper chromium and arsenic from industrial wood preservative sludge, Waste Manage. 22 (2002) 293301. D.C. Rice, Behavioral impairment produced by low level post natal PCB exposure in monkeys, Environ. Res. A 80 (1999) 113S121. M.R. Baloga, J.E. Earley, The kinetics of the oxidation of Cr(III) to Cr(VI) by hydrogen peroxide, J. Am. Chem. Soc. 83 (1961) 49064909. M. Pettine, F.J. Millero, Chromium speciation in seawater: the probable role of hydrogen peroxide, Limnol. Oceanogr. 35 (1990) 730736.